Health Services Research © Health Research and Educational Trust DOI: 10.1111/1475-6773.12139 HEALTH INFORMATION TECHNOLOGY

Where Are We on the Diffusion Curve? Trends and Drivers of Primary Care Physicians’ Use of Health Information Technology Anne-Marie Audet, David Squires, and Michelle M. Doty Objective. To describe trends in primary care physicians’ use of health information technology (HIT) between 2009 and 2012, examine practice characteristics associated with greater HIT capacity in 2012, and explore factors such as delivery system and payment reforms that may affect adoption and functionality. Data. We used data from the 2012 and 2009 Commonwealth Fund International Health Policy Surveys of Primary Care Physicians. The data were collected in both years by postal mail between March and July among a nationally representative sample of primary care physicians in the United States. Study Design. We compared primary care physicians’ HIT capacity in 2009 and 2012. We employed multivariable logistic regression to analyze whether participating in an integrated delivery system, sharing resources and support with other practices, and being eligible for financial incentives were associated with greater HIT capacity in 2012. Principal Findings. Primary care physicians’ HIT capacity has significantly expanded since 2009, although solo practices continue to lag. Practices that are part of an integrated delivery system or share resources with other practices have higher rates of electronic medical record (EMR) adoption, multifunctional HIT, electronic information exchange, and electronic access for patients. Receiving or being eligible for financial incentives is associated with greater adoption of EMRs and information exchange. Conclusions. Federal efforts to increase adoption have coincided with a rapid increase in HIT capacity. Delivery system and payment reforms and federally funded extension programs could offer promising pathways for further diffusion. Key Words. EMR adoption, primary care physicians, shared resources, financial incentives, health information exchange, small physician practices

National trends in physicians’ use of health information technology (HIT) are clear: in 2001, 18 percent of office-based physicians had implemented either a partial or full electronic medical record (EMR), 48 percent had done so in 347

348

HSR: Health Services Research 49:1, Part II (February 2014)

2009, and 72 percent in 2012 (Hsiao and Hing 2012). Less is known about the nature of the progress in areas such as computerized physician ordering or clinical decision support. Furthermore, little is known about how external factors such as shared technical assistance programs and delivery system and payment reforms are impacting adoption. Studies have described significant variation in EMR adoption by physician practice size, setting, and physician age (Decker, Jamoom, and Sisk 2012). Documented barriers include costs, the lack of expertise, and technical support infrastructure (Audet et al. 2004; DesRoches et al. 2008; Boonstra and Broekhuis 2010; Patel et al. 2013). The Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009 injected unprecedented resources to build the national HIT infrastructure. Those resources support several programs, including the “Meaningful Use” incentives, technical assistance through regional extension centers, and the creation of health information exchange communities. In addition, the Patient Protection and Affordable Care Act (ACA) of 2010 fostered payment and delivery system reforms, such as an increased focus on the patient-centered medical home model and accountable care organizations (ACOs). These initiatives aim to create and support an environment and infrastructure where all primary care practices can provide advanced primary care, regardless of their size or setting. In this study, we use results from two international surveys of primary care physicians—the first conducted in 2009, before the ACA and in the very early stages of deployment of HITECH funds, and the second in 2012—to describe trends in adoption of HIT functionalities. While HIT adoption in a cross-national context has been reported elsewhere (Schoen et al. 2009, 2012), this study further explores the factors that may facilitate or hinder HIT adoption and functionality in the United States.

DATA AND M ETHODS Data This study primarily uses data from the 2012 Commonwealth Fund International Health Policy Survey of Primary Care Physicians, conducted in 11 Address correspondence to Anne-Marie J. Audet, M.D., M.Sc., S.M., Delivery System Reform & Breakthrough Opportunities, The Commonwealth Fund, 1 East 75th Street, New York, NY 10021; e-mail: [email protected]. David Squires, M.A., is with the International Program in Health Policy and Practice Innovations, The Commonwealth Fund, New York, NY. Michelle M. Doty, Ph.D., is with the Survey Research and Evaluation, The Commonwealth Fund, New York, NY.

Trends and Drivers of HIT Use by Physicians

349

countries (Schoen et al. 2012), and restricts the analysis to U.S. respondents. The survey was conducted by Harris Interactive by postal mail between March and July; the response rate was 35 percent. The sample consisted of 1,012 primary care physicians—39 percent in family practice, 36 percent in internal medicine, 22 percent in pediatrics, and 3 percent in general practice —randomly drawn from the AMA Masterfile. Responses are weighted by gender, age, region, and specialty. We compared unweighted respondents and nonrespondents by region, years in practice, and type of practice (i.e., solo, group) and found similar proportions in both groups, suggesting that our sample is representative of the physicians in the AMA Masterfile. However, the low response rate does introduce potential bias. Data from the 2009 Commonwealth Fund International Health Policy Survey of Primary Care Physicians (Schoen et al. 2009) are also used to explore changes in HIT adoption over time. Both survey instruments are available in the online Appendix. The 2009 survey was conducted following the same methodology as in 2012; the response rate in 2009 was 39 percent. HIT Functionality Respondents were asked whether they use electronic patient medical records in their practice. They were also asked about several electronic HIT functions. Using the same methodology as Schoen et al. (2012), we grouped 15 HIT questions into four domains related to (1) generating patient information; (2) generating patient registry and panel information; (3) electronic order entry, and (4) electronic decision support (Table 1 includes a list of questions under each domain). Respondents who reported using electronic medical records and two or more functions within each of the four domains were considered to have “multifunctional” HIT capacity. Physicians with electronic exchange capacity reported having at least two of the following: the ability to electronically exchange clinical summaries or laboratory test results with doctors outside their practice, or to receive hospital discharge reports via remote access. Practices providing electronic access for patients offered patients two or more of the following: the option to request appointments or referrals online, email about a medical question, request online prescription refills, or view test results on a secure website. Integrated Delivery System and Sharing Resources We characterized respondents by whether they reported being part of an integrated system such as Kaiser or the V.A. In addition, the survey asked

350

HSR: Health Services Research 49:1, Part II (February 2014)

Table 1: Health Information Technology (HIT) Capacity among U.S. Primary Care Physicians 2009 and 2012 2009 (N = 1,442), % Uses electronic patient medical records Computerized ability to generate patient information List medications taken by an individual patient, including those prescribed by other doctors List lab results for an individual patient, including those ordered by other doctors Can provide patients with clinical summaries for each visit Can electronically generate patient information: 2 + of above 3 functions Computerized ability to generate patient registry and panel information List patients by diagnosis List patients due/overdue for tests or preventive care List all patients taking a particular medication List patients by lab result Can electronically generate patient registry and panel information: 2 + of above 4 functions Computerized order entry management: routinely… Order laboratory tests electronically Prescribe medication electronically Able to electronically send prescriptions to the pharmacy† Electronically track all laboratory tests until results reach clinicians Routinely uses computerized order entry management: 2 + of above 4 functions Computerized decision support: routinely… Receive electronic alert/prompt about potential drug dose/interaction problems Receive computerized reminder for guideline-based intervention/screening tests Receive computerized alert/prompt to provide patients with test results Send patients reminders for preventive/follow-up care through computerized system Routinely uses computerized decision support: At least 2 of the above 4 functions Multifunctional HIT capacity: Uses EMRs and at least two electronic functions for generating patient information, generating panel information, order entry management, and routine decision support Electronic information exchange Can electronically exchange clinical summaries with doctors outside practice

2012 (N = 1,012), %

46

69***

30

45*** 43 43 44

42 29 29

38 40 34 28

50*** 42*** 37 42*** 46

54*** 64*** 66*** 41*** 68

37

58***

20

33***

22

35***

18

22*** 45 27

33 continued

Trends and Drivers of HIT Use by Physicians

351

Table 1. Continued 2009 (N = 1,442), % Can electronically exchange laboratory/diagnostic tests with doctors outside practice Usually receive hospital discharge reports through remote access Able to electronically exchange information: 2 + of above 3 functions Electronic access for patients: Practice allows patients to… Request appointments or referrals online Email about a medical question or concern Request refills for prescriptions online View test results on a secure website Electronic patient engagement: 2 + of above 4 functions

2012 (N = 1,012), % 35 12 32 30 34 36 28 35

Statistically different: *p < .05, **p < .01; ***p < .001. Only asked of those who report occasionally or routinely electronically prescribing medication. Source. 2009 and 2012 Commonwealth Fund International Health Policy Surveys of Primary Care Physicians.



respondents whether they have formal arrangements with external organizations to share technical support for clinical information systems or quality improvement consultants or support. Those that reported either or both arrangements were characterized as sharing resources. Financial Incentives Our analysis characterized respondents by whether they received or had the potential to receive financial support for any of the following: managing patients with chronic disease or complex needs; providing enhanced preventive care activities; adding nonphysician staff to their practice team; having non–face-to-face interactions with patients; or making home visits. Practice Size Practice size was categorized into four groups: solo, 2 to fewer than 10 physicians, 10 to fewer than 20 physicians, and 20 or more physicians. Analysis Trends between 2009 and 2012 in the percentage of physicians reporting various HIT functions were examined (Table 1). Bivariate analyses and

352

HSR: Health Services Research 49:1, Part II (February 2014)

multivariable logistic regression of the 2012 data examined the association of HIT capacity with practice size, participation in an integrated delivery system, sharing resources, and eligibility for financial incentives, controlling for physician age, gender, urbanicity, and census region (Tables 2 Table 2: Practice Characteristics and HIT Capacity, U.S. Primary Care Physicians (2012) Unweighted N

Uses Electronic Patient Medical Records

Multifunctional HIT Capacity†

Can Electronically Exchange Information with Other Providers‡

U.S. 1,012 69% 27% 32% total Formal arrangements with other practices/groups to share technical support for clinical information systems, and/or quality improvement consultants or support Yes 376 83*** 38*** 49*** No 580 59 20 21 Part of integrated system (e.g., Kaiser, the V.A.) Yes 276 87*** 41*** 51*** No 721 63 22 24 Receives or eligible for targeted financial incentives¶ Yes 342 76*** 31* 39*** No 653 65 25 27 Practice size†† Solo 238 49 11 19 2 to 485 70*** 29*** 32***

Where are we on the diffusion curve? Trends and drivers of primary care physicians' use of health information technology.

To describe trends in primary care physicians' use of health information technology (HIT) between 2009 and 2012, examine practice characteristics asso...
142KB Sizes 0 Downloads 0 Views